Enhancements to layer 2 (L2) protocols, (i.e., medium access control (MAC) and radio link control (RLC)), for the high speed uplink packet access (HSUPA) have been introduced into the third generation partnership project (3GPP) Release 8. These enhancements allow better support of high data rates by introducing flexible RLC protocol data unit (PDU) sizes along with segmentation capabilities at the MAC layer of the wireless transmit/receive unit (WTRU). MAC-i/is sub-layer, (MAC entity for handling enhanced dedicated channel (E-DCH)), includes a segmentation entity per logical channel identity. If a dedicated channel MAC (MAC-d) PDU or a common channel MAC (MAC-c) PDU does not fit into the available transport block space, the MAC-i/is entity segments the MAC-d or MAC-c PDU and stores the remaining segment in the segmentation entity.
Since the introduction of the HSUPA in the 3GPP Release 6, mechanisms have been introduced to help the Node B scheduler assess the uplink data rate that the WTRU requires and assist the Node B to get a better view of the amount of system resources needed by the WTRU and the amount of resources the Node B can actually make use of. One of these mechanisms is scheduling information (SI) that may be included into a MAC-e or MAC-i PDU. The scheduling information includes a total E-DCH buffer status (TEBS) field which identifies the amount of data in number of bytes that is available for transmission and retransmission in the RLC layer. The scheduling information also indicates a highest priority logical channel (HLID) with available data, the highest priority logical channel buffer status (HLBS), and the UE power headroom (UPH). The total length of the scheduling information is 18 bits for frequency division duplex (FDD). For time division duplex (TDD) the scheduling information further includes a field to assist the Node B with estimation of the degree of inter-cell interference a WTRU will generate.
The scheduling information reporting is triggered according to a number of different triggering mechanisms that are dependent on the value of the serving grant and the TEBS. The scheduling information reporting is triggered if the serving grant is equal to “zero” or all HARQ processes are deactivated, and the TEBS becomes larger than zero. The scheduling information reporting may be triggered periodically, which is configured by RRC. Two different periodic triggers, (i.e., timers), may be configured for two cases. In the case where the serving grant is zero or all processes are deactivated, and TEBS is larger than zero, T_SING is the timer that determines the periodicity of the triggering. In the case where the serving grant is not zero and at least one process is activated, T_SIG is the timer that determines the periodicity of the triggering. The scheduling information reporting may be triggered if the serving grant becomes too small to allow transmission of a single PDU from any scheduled MAC-d flow and TEBS becomes larger than zero. The scheduling information reporting may be triggered if an E-DCH serving cell change occurs and the new E-DCH serving cell was not part of the previous serving E-DCH radio link set (RLS). The scheduling information is not transmitted if TEBS is zero, even if it is triggered by one of the configured triggering mechanisms.
A happy bit is an additional mechanism that allows the Node B to know whether the WTRU is happy with the current grant and resources allocated to the WTRU. The happy bit is a single bit field that is passed from the MAC entity to the physical layer for inclusion on an E-DCH dedicated physical control channel (E-DPCCH). The happy bit may be set to either “happy” or “not happy.” “Happy” indicates that the WTRU could use more resources, and “not happy” indicates that the WTRU could not use more resources. The setting of the happy bit is also based on the TEBS.
One of the conditions to set the happy bit to “not happy” is if the TEBS would require more than Happy_bit_delay_condition ms to be transmitted with the current serving grant×the ratio of active HARQ processes to the total number of HARQ processes.
According to the current 3GPP specifications, the TEBS sent as part of the scheduling information only indicates the total amount of data in bytes that is available across all logical channels that are available for transmission and retransmission in the RLC layer. However, with the introduction of segmentation capabilities in the MAC entity, (i.e., MAC-i/is entity), the MAC entity may have one or more segments stored in the segmentation entity. The total number of bytes waiting in the MAC-i/is segmentation entity is not included in the TEBS calculation. Due to the fact that the introduction of flexible RLC PDUs allows the RLC entity to create RLC PDUs of a size up to the maximum RLC PDU size, the number of bits stored in the segmentation entity could be relatively high. Therefore, by not including the amount of data stored in the MAC layer segmentation entity, the WTRU reports an incorrect TEBS. Since the WTRU requires additional resources to transmit the segments stored in the MAC layer segmentation entity, by not reporting them, the network will obtain an incorrect view of the number of resources required by the WTRU.
Additionally, where the network decreases the grant of the WTRU to zero the WTRU may still have some data in the MAC layer segmentation entity to transmit. The current scheduling information triggering mechanisms allow the WTRU to send the scheduling information if the TEBS becomes larger than zero and the serving grant is zero. However, if TEBS is zero but the WTRU has some remaining data in the MAC layer segmentation entity, the WTRU transmission will stall in that situation. The triggering criteria will not allow the WTRU to send a TEBS and thus the network will not know that the WTRU requires additional resources to successfully complete the transmission of the data in its buffer. This will delay transmission and prevent the WTRU from completing transmission of the data.
The setting of the happy bit is also dependent on the TEBS. If TEBS is zero or if TEBS is incorrectly set and the WTRU has one or more MAC segments in the MAC layer segmentation entity, the WTRU will not set the happy bit properly. Due to the fact that the TEBS is underestimated, the WTRU might calculate that the WTRU is able to transmit the data within the delay restrictions. However, if the data contained in the MAC layer was taken into consideration the WTRU may not be able to satisfy the delay criteria.